Falsifiable implies Learnable
arXiv:1408.6618v1
Originality Incremental advance
AI Analysis
This work addresses a foundational problem in machine learning theory by linking falsifiability to learnability, which is incremental as it builds on existing theoretical frameworks.
The paper proves that falsifiability is a necessary condition for learnability in statistical learning and sequential prediction, showing that if a theory is falsifiable, it admits an optimal prediction strategy, with analogous results for universal induction.
The paper demonstrates that falsifiability is fundamental to learning. We prove the following theorem for statistical learning and sequential prediction: If a theory is falsifiable then it is learnable -- i.e. admits a strategy that predicts optimally. An analogous result is shown for universal induction.